Skip to content

originaonxi/ARIA

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

16 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

ARIA — Autonomous Relationship Intelligence Agent

One of two permanent intelligence cores at Aonxi. ARIA finds capital. AROS finds revenue. Both learn from the same brain. Both compound forever.

Status PKM Brain Raising

Read the full Aonxi vision


What ARIA does

ARIA finds investors, scores them by thesis fit, verifies every email, profiles their psychological defense mode, writes a message calibrated to bypass that defense, creates the campaign, and sends it.

Autonomously. The same system that finds revenue now finds capital.

First deployment — March 2026:

Investors contacted:  General Catalyst, Bessemer, Antler,
                      Blume VC, Kae Capital + 14 more
Emails sent:          19
Bounces caught:       4 (before sending — domain reputation protected)
Raising:              $250K pre-seed
Product:              $199K collected, $0 raised, 40 customers

The product being used to raise money is being used to raise the money. Every investor who receives an ARIA email is watching it work before they reply.


Why VCs are the hardest defense mode to bypass

VCs are MOTIVE_INFERENCE — awareness score 9/10.

They spend their careers evaluating pitches. They have read 10,000 cold emails. They decoded your intent before you finished the subject line.

Most founder emails look like this:

Hi Sarah — I'm really excited to share what we're building at Aonxi. I think there's a great opportunity here and would love to connect for a quick call.

Every word triggers a defense. Excited — detected. Great opportunity — detected. Quick call — detected. Email deleted before the second paragraph.

ARIA sends emails that look like this:

$199K collected. $8K peak day. $0.50/day to run. Zero sales reps. Code is public. You backed [company] — autonomous GTM at the SMB layer. This is that, built and verified. 20 minutes. No deck.

No "I". No pitch language. No excitement. Data, a specific signal, a specific ask.

That is PURE_DATA bypass for MOTIVE_INFERENCE defense.


The PKM layer in investor outreach

Same 10 defense modes. Different population distribution. PKM Analyzertry it live (free, no key needed).

Investor type Defense mode Bypass What changes
VC partner (ex-GTM) MOTIVE_INFERENCE PURE_DATA Opens with a number, never "I'm excited to"
Operator angel TACTIC_RECOGNITION SIGNAL_HOOK References their specific portfolio company
Technical investor SOCIAL_PROOF_SKEPTICISM CREDIBILITY_FIRST Verifiable numbers only, no "trusted by"
First-time angel AUTHORITY_DEFERENCE PEER_PROOF Names other angels who committed
Busy SMB operator OVERLOAD_AVOIDANCE ULTRA_SHORT 60-word hard cap, specific calendar slot

ARIA detects which one. ARIA writes for that one. Every time.

All defense profiles are cached in Airtable — shared across ARIA and AROS. A prospect analyzed by one agent is instantly available to the other.

Based on Friestad & Wright (1994) — 30 years of persuasion psychology.


The brain — Airtable

Every investor ARIA contacts is recorded.

  • Which thesis keywords converted
  • Which defense mode each investor was running
  • Which bypass strategy got the reply
  • Which time of day a specific fund partner opens emails
  • Which portfolio signal triggered the conversation

This pattern library compounds every raise. The next founder who uses ARIA gets the benefit of every raise that came before them.


Architecture

aria.py               CLI entry — 13 commands, one pipeline
aria_db.py            SQLite dedup engine (source of truth)
apollo_client.py      Investor finding + enrichment
millionverifier.py    Email verification before send
investor_scorer.py    Thesis fit scoring (Tier 1/2/3)
investor_researcher.py SerpAPI research
investor_writer.py    Claude Haiku email generation (PKM applied)
instantly_client.py   Campaign creation + send + tracking
airtable_sync.py      Visual CRM dashboard sync
linkedin_prep.py      HeyReach CSV for LinkedIn outreach
reply_processor.py    Claude reply classification + HOT alerts
briefing.py           Daily 7am pipeline email

One command:

python aria.py run

Find → Score → Verify → Research → Profile → Write → Send → Track → Learn.

Stack: Claude Haiku / Apollo / Millionverifier / SerpAPI / Instantly / Airtable / SQLite


The 5-generation roadmap

Gen Name Status What it adds
1 Pipeline LIVE Find → Score → Verify → Research → Profile → Write → Send
2 Learner Week 4 A/B testing, auto-adjust scoring from reply data
3 Researcher Week 8 Real-time portfolio moves, trigger-based outreach
4 Negotiator Week 12 Objection handling, warm intro path mapping
5 Autonomous Week 16 Zero human input, self-sourcing, self-qualifying

All commands

Command What it does
aria run Full pipeline: find → verify → score → research → write → send
aria run --auto Same but skip preview — send all automatically
aria status Pipeline dashboard
aria replies Check Instantly for replies, classify, alert on HOT
aria followup Day 5-8 follow-ups for non-responders
aria linkedin Generate HeyReach CSV for LinkedIn outreach
aria briefing Send daily pipeline email
aria load FILE Import CSV manually
aria verify Run Millionverifier on unverified emails
aria score Score all unscored investors
aria research N Research top N investors via SerpAPI
aria write Write emails for top scored investors
aria sync Sync to Airtable CRM dashboard

Run locally

git clone https://github.com/originaonxi/ARIA
cd ARIA
pip install -r requirements.txt
cp .env.example .env
# Add your keys to .env
python aria.py run

Related

Repo What it does
AROS Finds revenue — the other intelligence core
PKM Analyzer Defense profiling — try it live
VISION.md The full Aonxi vision — AGI revenue layer for 400M businesses

Built by Anmol Samorigin@aonxi.com / originaonxi.github.io / aonxi.app

About

Autonomous investor outreach agent — finds investors, scores by thesis fit, writes 3-sentence emails via Claude, verifies deliverability, sends via Instantly. One command: aria load investors.csv

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages